suppress pyre erros in xnnpack/fuse_batch_norm

Summary:
There were multiple pyre errors in fuse_batch_norm operator before this diff.

 This diff added several pyre command to suppress them.

Before:
 {F1128589876}

After:
 {F1128590200}

Reviewed By: digantdesai

Differential Revision: D50563193

fbshipit-source-id: 5843d32e9b8724399c4aaedbacd31b918d8d5748

backends/xnnpack/passes/fuse_batch_norm_with_conv.py

@@ -66,7 +66,10 @@ def call(self, graph_module: torch.fx.GraphModule):
 
             # Get the parameters from conv op
             assert len(conv.args) == 9
+
             conv_weight = get_param_tensor(self.exported_program, conv.args[1])
+            assert conv_weight is not None
+
             conv_bias = get_param_tensor(self.exported_program, conv.args[2])
 
             # Get the parameters from the batchnorm op
@@ -80,8 +83,12 @@ def call(self, graph_module: torch.fx.GraphModule):
             )
             bn_weight = get_param_tensor(self.exported_program, bn.args[1])
             bn_bias = get_param_tensor(self.exported_program, bn.args[2])
+
             running_mean = get_param_tensor(self.exported_program, bn.args[3])
+            assert running_mean is not None
+
             running_var = get_param_tensor(self.exported_program, bn.args[4])
+            assert running_var is not None
 
             # args[7] for native_batch_norm, but args[6] for
             # _native_batch_norm_legit_no_training (which doesn't have training